Electrocardiography during exercise traditionally has been used to help determine the presence or absence of coronary artery disease in a given individual. Our results demonstrate that the results of such testing are relatively insensitive in patients with symptoms of coronary disease, and are very poor predictors of coronary artery disease in subjects without symptoms.